Package ‘ndtv’May 22, 2019

Type Package

Title Network Dynamic Temporal Visualizations

Version 0.13.0

Date 2019-05-21

Depends R (>= 3.0), network (>= 1.13),networkDynamic (>=0.9),animation (>= 2.4),sna

Imports MASS, statnet.common, jsonlite, base64

Suggests tergm (>= 3.6), tsna, testthat, knitr, htmlwidgets,scatterplot3d

Description Renders dynamic network data from 'networkDynamic' objects as movies, interactive ani-mations, or other representations of changing relational structures and attributes.

License GPL-3 + file LICENSE

URL https://github.com/statnet/ndtv

VignetteBuilder knitr

NeedsCompilation no

Author Skye Bender-deMoll [cre, aut],Martina Morris [ctb]

Maintainer Skye Bender-deMoll <skyebend@uw.edu>

Repository CRAN

Date/Publication 2019-05-22 07:20:03 UTC

R topics documented:ndtv-package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2compute.animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3effectFun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6export.dot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7export.pajek.net . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9filmstrip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10install.ffmpeg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1

2 ndtv-package

install.graphviz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12layout.center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13layout.distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14msm.sim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15ndtvAnimationWidget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17network.layout.animate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18proximity.timeline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21render.animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25render.d3movie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29stergm.sim.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34timeline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35timePrism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37toy_epi_sim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40transmissionTimeline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Index 45

ndtv-package Network Dynamic Temporal Visualization (ndtv)

Description

Construct visualizations such as timelines and animated movies of networkDynamic objects to showchanges in structure and attributes over time.

Details

For version and license information, run packageDescription('ndtv')

Uses network objects with dynamics encoded using networkDynamic.

Key features:

• Compute a dynamic layout using compute.animation.

• Render it as a movie using render.animation.

• Render animations to a web page using render.d3movie

• Plot multiple ’stills’ of a movie with filmstrip

• Plot a timeline of edge and vertex activity with timeline

• Plot network geodesic proximities as a stream graph with proximity.timeline

To view package vignettes and extended examples, see browseVignettes(package='ndtv').

For a more indpeth tutorial, see http://statnet.csde.washington.edu/workshops/SUNBELT/current/ndtv/ndtv_workshop.html

The package includes several example datasets:

• msm.sim output of a stergm simulation of basic sex contact network model

• short.stergm.sim Very Very Basic stergm simulation output (25 time steps)

compute.animation 3

• stergm.sim.1 Very Very Basic stergm simulation output (100 time steps)

• toy_epi_sim Toy Epidemic Simulation Output from the EpiModel package

Report bugs at: https://github.com/statnet/ndtv/issues

Author(s)

Skye Bender-deMoll, and the Statnet Commons team Maintainer: Skye <skyebend@uw.edu>

References

Bender-deMoll, S., Morris, M. and Moody, J. (2008) Prototype Packages for Managing and Ani-mating Longitudinal Network Data: dynamicnetwork and rSoNIA Journal of Statistical Software24:7.

Hunter DR, Handcock MS, Butts CT, Goodreau SM, Morris M (2008b). ergm: A Package to Fit,Simulate and Diagnose Exponential-Family Models for Networks. Journal of Statistical Software,24(3). http://www.jstatsoft.org/v24/i03/.

Butts CT (2008). network: A Package for Managing Relational Data in R. Journal of StatisticalSoftware, 24(2). http://www.jstatsoft.org/v24/i02/.

Carter T. Butts, Ayn Leslie-Cook, Pavel N. Krivitsky and Skye Bender-deMoll (2015). networkDy-namic: Dynamic Extensions for Network Objects. R package version 0.7. http://statnet.org

Skye Bender-deMoll and McFarland, Daniel A. (2006) The Art and Science of Dynamic NetworkVisualization. Journal of Social Structure. Volume 7, Number 2 http://www.cmu.edu/joss/content/articles/volume7/deMollMcFarland/

See Also

networkDynamic, compute.animation,render.animation for examples, and the package vignettevignette(package='ndtv').

compute.animation Compute a sequence of vertex layouts over time suitable for renderingan animation.

Description

Steps through a networkDynamic object and applies layout algorithms at specified intervals, storingthe calculated coordinates in the network for later use by the render.animation function. Generallythe layout are done in a sequence with each using the previously calculated positions as initial seedcoordinates in order to smooth out the resulting movie. Not all network layout algorithms give goodresults.

4 compute.animation

Usage

compute.animation(net, slice.par = NULL, animation.mode = "kamadakawai",seed.coords = NULL, layout.par = list(),default.dist = NULL, weight.attr = NULL, weight.dist=FALSE,chain.direction=c('forward','reverse'),verbose = TRUE,...)

Arguments

net A networkDynamic network object describing the temporal evolution of a net-work.

slice.par A list of parameters which specify the time steps and aggregation that should beused when moving through the network. Example: slice.par=list(start=0,end=100,interval=1, aggregate.dur=1,rule='latest')

The parameters are:

• start The time point at which the sequence of layouts should begin• end The time point at which the sequence of layouts should finish• interval The amount of time between successive layouts• aggregate.dur The duration of time over which the network should be

aggregated to derive the network for each layout• rule The aggregation rule to be used when collapsing the network.

animation.mode The name of the network animation layout to be used. These layouts are namenetwork.layout.animate.something but will be matched using the final part ofthe name. Current useful values are:

• network.layout.animate.kamadakawai essentially wrapper for the Ka-madakawai layout included in the network package.

• network.layout.animate.MDSJ a wrapper to do a Stress Majorization op-timized MDS layout using the Multi Dimensional Scaling for Java package.Note, due to license restrictions, this algorithm is for non-commercial useonly)

• network.layout.animate.useAttribute applies coordinates stored in auser-generated dynamic network attribute

• network.layout.animate.Graphviz a wrapper for the Graphviz softwarelibrary –if the library is installed on your system.

seed.coords (optional) an array of initial positions to be used for the very first layout in thesequence

layout.par A list of parameters to be passed to the layout algorithm.

default.dist The default distance to be used to separate nodes (or disconnected network com-ponents). Default to sqrt(network.size(net)). See layout.distance.

weight.attr charater providing the name of a (possibly dynamic) numeric edge attributedefining weights for the edges in each time slice. The values activity.durationor activity.count can be used to weight edges by the duration or count of theedge’s activity spells in the time slice.

weight.dist logical, defaults to FALSE, meaning that the edge weight values provided byweight.attr will be treated as similarities (larger values means closer). A value

compute.animation 5

of TRUE means that weights should be intrepreted as distances. See layout.distancefor more information.

chain.direction

a value of 'forward' indicates the chain of layouts should be computes in for-ward temporal order. A value 'reverse' runs the chain backwards. For somelayouts, reverse-chaining means that isolated vertices are more likely to havepositions close to the partners they will be tied to.

verbose If true, additional information about the layout process and progress will bereturned to console.

... possible additional arguments to be passed to sub processes

Details

This function is under active development so implementation and parameters will continue tochange.

Value

Invisibly returns original network argument (which is also modified in-place), with the additionof a network variable slice.par storing the slice parameters used, and dynamic node attributesanimation.x and animation.y storing the coordinates calculated for each time point.

Author(s)

Skye Bender-deMoll, and the statnet team.

References

See docs for specific layout functions.

Bender-deMoll, S., Morris, M. and Moody, J. (2008) Prototype Packages for Managing and Ani-mating Longitudinal Network Data: dynamicnetwork and rSoNIA Journal of Statistical Software24:7.

Krivitsky P and Handcock M (2012). Fit, Simulate and Diagnose Models for Network Evoluationbased on Exponential-Family Random Graph Models. Version 3.0-999. Project home page athttp://www.statnet.org, CRAN.R-project.org/package=tergm.

Butts CT (2008). network: A Package for Managing Relational Data in R. Journal of StatisticalSoftware, 24(2). http://www.jstatsoft.org/v24/i02/.

Skye Bender-deMoll and McFarland, Daniel A. (2006) The Art and Science of Dynamic NetworkVisualization. Journal of Social Structure. Volume 7, Number 2 http://www.cmu.edu/joss/content/articles/volume7/deMollMcFarland/

See Also

See also layout.distance, render.animation, network.layout.animate.MDSJ,ndtv, packagevignette (vignette('ndtv')) for examples.

6 effectFun

effectFun functions to manipulate graphic attributes of network for ’special ef-fects’

Description

Functions that can return appropriate graphic attributes (i.e. color interpolation) based on propertiesof the network (ages of edges,etc)

Usage

effectFun(name, ...)effect.edgeAgeColor (net, onset, fade.dur,

start.color = "#000000FF", end.color = "#00000000",na.color = "#CCCCCC55")

effect.vertexAgeColor (net, onset, fade.dur, start.color = "#000000FF",end.color = "#00000000", na.color = "#CCCCCC55")

Arguments

name the short name of the effect function to be returned. i.e 'edgeAgeColor' willreturn the effect.edgeAgeColor function

... additional arguments to be passed in to effect functions

net a network object to be evaluated

onset the time at which the network should be evaluated

fade.dur (effect property) numeric value giving the color duration of the interpolation

start.color (effect property) color name for color value to be used at start of interpolation

end.color (effect property) color name for color value to be used at start of interpolation

na.color (effect property) default color name for color value to be used for edge/verticesthat are not currently active

Details

The special effects functions can be called directly for use as graphic parameters with standardnetwork plots, or via effectFun which will return the effect in a functional form so that it canevaluated/substituted at each time point as plot control function to render.animation

effect.edgeAgeColor calculates the edge of each edge in net at the time onset and uses the valueto return a color interpolated between start.color and end.color by comparing the time of eachedge to the fade.dur parameter

effect.vertexAgeColor does the same, but for vertices.

Users can also define functions to be called in this way

export.dot 7

Value

effectFun returns the function named by its first argument, with any arguments matching in ...substituted.

Author(s)

skyebend@uw.edu

Examples

library(ndtv)data('short.stergm.sim')# render a plot with edges colored by age at time 24# edges labeld with ageplot(short.stergm.sim,edge.col=effect.edgeAgeColor(short.stergm.sim,

fade.dur=25,start.color = 'red',end.color='blue',onset=24),

edge.label=edges.age.at(short.stergm.sim,24),edge.lwd=5)

## Not run:# render an animation where edges are colored dynamically by their age# starting out red and fading to bluecompute.animation(short.stergm.sim,slice.par = list(start=0,

end=25,interval=1, aggregate.dur=5,rule='latest'))

render.animation(short.stergm.sim,edge.col=effectFun('edgeAgeColor',fade.dur=5,start.color = 'red',end.color='blue'))

## End(Not run)

export.dot Export a network file as Graphviz .dot formatted text file.

Description

A crude exporter for saving out a network in the Graphviz .dot format. http://graphviz.org/content/dot-language

Usage

export.dot(x, file = "", coords = NULL, all.dyads = FALSE,vert.attrs = NULL, edge.attrs = NULL)

8 export.dot

Arguments

x The network object to be exported

file The file name where network should be saved

coords Optional node coordinates to include

all.dyads FALSE, a numeric value, or a symetric matrix of distances providing the desiredlengths for all dyads.. If numeric, entries are written out for all possible dyadsin the network, and the numeric value will be used to fill in the values for allthe dyads in the matrix not linked by an edge (see default.dist param tolayout.distance). This is necessary for some uses cases, but will generatedramatically larger files and slower performance. For the matrix and numericcases, the values will be written as Graphviz ’len’ edge attributes, and the valuesof edge.attrs will be ignored.

vert.attrs optional character vector listing the names of any vertex attributes of the networkthat should be included as attributes of the nodes in the Graphviz dot file. (e.g.’label’, ’width’)

edge.attrs optional character vector listing the names of any edge attributes of the networkthat should be included as attributes of edges in the Graphviz dot file. (e.g.’weight’,’penwidth’)

Details

A crude exporter for saving out a network in the Graphviz .dot format. http://graphviz.org/content/dot-language

Value

Returns nothing but creates a file in .dot format: http://graphviz.org/content/dot-language

Note

This is still a partial implementation focusing on edges, edge wights, and node coordinates in orderto pass the information to graphViz to use it as an external layout engine rather than a renderer.

Author(s)

Skye Bender-deMoll

References

http://graphviz.org/content/dot-language

Examples

library(network)net <- network.initialize(5)net[1,] <-1net[2,3] <-2export.dot(net,file="testNet.dot")

export.pajek.net 9

export.pajek.net Export a network file as a Pajek .net formatted text file.

Description

A basic tool for exporting a network as a Pajek http://pajek.imfm.si/doku.php?id=pajek .netformat text file. Does not yet encode attributes, layout information or timing info.

Usage

export.pajek.net(net, filename)

Arguments

net a network objectfilename the file where the network object should be saved

Details

.net is basically an edgelist format with sections for vertices, arcs and edges. Vertex attributes for’label’, coordinates named ’x’,’y’,’z’, ’color’ as ’ic’ (inner color), ’shape’ as a shape value will bewritten in the appropriate Pajek format. An edge attribute of ’weight’ will be written as the edgevalue, ’width’ as ’w’ and ’color’ as ’c’. See read.paj for reading pajek files (time info supported)

Value

A file is written out containing the vertex and edge data.

Note

This is a very minimal implementation, mostly used for testing layout algorithms. Timing informa-tion is not yet supported.

Author(s)

Skye Bender-deMoll

References

Pajek software: http://pajek.imfm.si/doku.php?id=pajek

Pajek file format documentation: http://vlado.fmf.uni-lj.si/pub/networks/pajek/svganim/1.10.7.1/PajekToSvgAnim.pdf

Examples

data('toy_epi_sim')toy_epi_sim%v%'color'<-'blue'export.pajek.net(toy_epi_sim,filename='toy_epi_sim.net')

10 filmstrip

filmstrip Create a ‘small multiples‘ plot of a networkDynamic object.

Description

Plots several frames of a network animation of a networkDynamic object in a single static image asa way to provide a quick visual summary of the dynamics of the network.

Usage

filmstrip(nd, frames = 9, slice.par, render.par, mfrow, verbose = FALSE, ...)

Arguments

nd networkDynamic object to be plotted

frames integer number of frames to extract and render

slice.par optional list of parameters to control binning of network, overrides frames ar-gument. See compute.animation

render.par optional list of parameters to control rendering of network. See render.animation

mfrow optional two-element numeric vector giving the number of rows and columnsfor the layout grid. See par.

verbose boolean,(defaults to FALSE) verbose argument to be passed to compute.animationrender.animation/

... additional arguments to be passed to plot.network via render.animation

Details

If the networkDynamic object does not already have animation coordinates, calls compute.animationto calculate coordinates for the appropriate number of frames. The frames argument determinesthe number of evenly-spaced network slices to be rendered by render.animation (with the nor-mal plot recording disabled) as images on the final plot grid. Note that if the layout has coordinatespre-computed by compute.animation, the slices selected by the frames argument may not alignexactly with the previously compute slices. Passing in a slice.par argument will overide framesto determine exactly which slices will be rendered.

The layout of plot grid can be changed via the mfrow argument.

Value

Generates plots on the active graphics device.

Note

This is a DRAFT version of the function.

Author(s)

skyebend

install.ffmpeg 11

See Also

See also compute.animation, render.animation.

Examples

data(stergm.sim.1)filmstrip(stergm.sim.1,displaylabels=FALSE)# print an overall title for the main plottitle('stergm.sim.1 at 9 time points')

# adjust margins of individual plots to make more roompar(mar=c(1,1,1,1))filmstrip(stergm.sim.1)

install.ffmpeg Instructions for installing ffmpeg on various platforms

Description

The animation package uses ffmpeg to export movies into video formats. This internal functiondoesn’t actually install the ffmpeg library, it just gives instructions on how to do the installation –which really just point to these docs.

Usage

install.ffmpeg()

Details

Here are some all-too-brief instructions for the various platforms. After you have installed FFmpegon your system, you can verify that R knows where to find it by typing Sys.which('ffmpeg') inthe R terminal. You many need to first restart R after the install.

Installing in Windows:• Download the recent ’static’ build from http://ffmpeg.zeranoe.com/builds/

• Downloads are compressed with 7zip, so you may need to first install a 7zip decompressionprogram before you can unpack the installer.

• Decompress the package and store contents on your computer (probably in Program Files)• Edit your system path variable to include the path to the directory containing ffmpeg.exe

Installing on a Mac:• Download most recent build from http://www.evermeet.cx/ffmpeg/• The binary files are compressed with 7zip so may need to install an unarchiving utility: http://wakaba.c3.cx/s/apps/unarchiver.html

• Copy ffmpeg to /usr/local/bin/ffmpeg

Installing in Linux/Unix (ffmpeg or avconv):

12 install.graphviz

• FFmpeg is a standard package on many linux systems. You can check if it is installed with acommand like dpkg -s ffmpeg. If it is not installed, you should be able to install with yoursystem’s package manager. i.e. sudo apt-get install ffmpeg or search ’ffmpeg’ in theSoftware Center on Ubuntu.

• Ubuntu and Debian systems may use an alternate program named "avconv" which can be in-stalled with sudo apt-get install libav-tools or by searching ’libav-tools’ in Ubuntu’sSoftware Center. Verify that R knows where to find it by typing ‘Sys.which(’avconv’)‘ in theR terminal. You many need to first restart R after the install. The animation library shouldautomatically use ’avconv’ if it sees it instead of ’ffmpeg’. If it doesn’t, you can tell it to bytyping ani.options(ffmpeg='avconv') in your R session

Value

On winddows: Will open a web browser window to the ffmpeg website and give instructions howto open this help file.

References

http://ffmpeg.org

install.graphviz Instructions for installing the Graphviz libraries on various platforms

Description

The network.layout.animate.Graphviz layout provides an interface for calling the various lay-outs provided by the Graphviz library (http://www.graphviz.org) if it is installed on your sys-tem. Since Graphviz is not an R package, you must manually install it on your system to get it towork.

Usage

install.graphviz()

Details

This function doesn’t actually install Graphviz, it just points to these docs which give a very briefoverview of how to do it on each platform.

Installing on Windows:• download the "current stable release" installer from https://graphviz.gitlab.io/_pages/Download/Download_windows.html

• run the installer• Edit your system path variable to include the path to the directory containing the graphviz

.exe files.

Installing on a Mac: It seems that there is no longer a .pkg for the mac, but it can be installedeasily via homebrew

layout.center 13

• install the brew package manager from https://brew.sh/

• from the Terminal, run brew install graphviz

Installing in Linux/unix:

• Graphviz is a standard package on many linux distributions. You can check if it is installedwith a command like dpkg -s graphivz. If it is not installed, you should be able to installit with your system’s package manager. i.e. sudo apt-get install graphviz or search’graphivz’ in the Software Center on Ubuntu.

When Graphviz is installed correctly on any platform the R command Sys.which('neato') shouldprint out the path to the installed libraries.

Value

On some platforms this function will open a web browser pointing to the download page forGraphviz.

Author(s)

skyebend

References

John Ellson et.al (2001) "Graphviz – open source graph drawing tools" Lecture Notes in ComputerScience. Springer-Verlag. p483-484 http://www.graphviz.org

See Also

See network.layout.animate.Graphviz for more details about how ndtv usees Graphviz.

layout.center Functions to center and normalize the coordinates of a network plotwithin a window.

Description

The layout.center function takes a matrix of coordinates and an x- and y-coordinate range andcenters the input coordinates within the range.

The layout.normalize function takes a matrix of coordinates and rescales them to the range (-1,1).If keep.aspect.ratio=FALSE, x- and y-coords are rescaled independently.

Usage

layout.center(coords, xlim, ylim)layout.normalize(coords, keep.aspect.ratio = TRUE)

14 layout.distance

Arguments

coords two column numeric matrix of coordinates.

xlim two element numeric vector giving min and max of x axis

ylim two element numeric vector giving min and max of y axiskeep.aspect.ratio

boolean, if FALSE, x- and y-axis will be rescaled indpendently

Details

These functions are used internally, but can also be called by the user when manipulating coordi-nates for layouts, especially when the coordinate ranges for a sequence of layouts do not match upwell. TODO: add barycenter function, and center on vertex function

Value

The input two column numeric matrix of coordinates with positions transformed.

Author(s)

skyebend

Examples

data(McFarland_cls33_10_16_96)coords<-plot(cls33_10_16_96)

# center layout coords with 100 unit arealayout.center(coords,xlim=c(0,100),ylim=c(0,100))

# rescale layout coords to unit intervallayout.normalize(coords)

layout.distance Provides a default way to convert a network into a set of euclidiandistances suitable for MDS-style layout optimization.

Description

Computes a geodesic path distance matrix for a network after symmetrizing, replacing Inf valueswith default.dist

Usage

layout.distance(net, default.dist = NULL, weight.attr = NULL,weight.dist = FALSE)

msm.sim 15

Arguments

net The network that the distance matrix should be computed fordefault.dist An (optional) value to be used to replace undefined values created by isolates

and disconnected components.weight.attr character, (optional) the name of an edge attribute of net containing numeric

values to use for edge distances.weight.dist logical, should the edge values given by weight.attr be interpreted as dis-

tances (larger values should place vertices farther apart) ? Default (FALSE) as-sumes values are similarities (larger values means stronger connection meansvertices closer together ).

Details

If no default.dist is provided the value sqrt(network.size(net)) will be used. If input issimilarity, it will be recoded/reversed to distances by subtracting each non-zero value from the maxvalue of the matrix and adding the min value of the matrix. If the network is directed, the matrixwill then be symmatrized to either the max value of i-j relation (if weight.dist=FALSE) or minvalue of i-j relation (if weight.dist=TRUE). Note that if the network is marked as undirected butincludes bi-directional edges, the (i,j) value will be chosen instead of (j,i).

Value

A distance matrix assumed to be appropriate for the network.

Author(s)

Skye Bender-deMoll

Examples

test<-network.initialize(4)add.edges(test,tail=1:2,head=2:3)# in adjacency matrix formas.matrix(test)# as matrix of geodesic distanceslayout.distance(test,1.5)

msm.sim MSM.sim : output of a stergm simulation of basic sex contact networkmodel

Description

A 1000 vertex networkDynamic object that contains the output of 10 timesteps of a discrete stergmsimulation of a basic sex contact network model. The model has two vertex types (’races’) withdifferent contact preferencs expressed with a nodematch parameter. The output network object isincluded as an example because re-running the model can take a while.

16 msm.sim

Usage

data(msm.sim)

Format

a networkDynamic object

Details

The model was built with the following stergm:

msm.net <- network.initialize(1000, directed=F)msm.net %v% 'race' <- rep(c(1,2),500)sm.form.constraints <- ~bd(maxout=2)msm.form.formula <- ~edges+nodematch('race')+degree(2)msm.target.stats <- c(450,375,50)msm.diss.formula <- ~offset(edges)+offset(nodematch("race"))msm.theta.diss <- c(2.944, -0.747)msm.fit <- stergm(msm.net,formation= msm.form.formula,dissolution= msm.diss.formula,targets="formation",target.stats= msm.target.stats,offset.coef.diss = msm.theta.diss,form.constraints=msm.form.constraints,estimate = "EGMME",control=control.stergm(SA.plot.progress=TRUE))msm.sim <- simulate(msm.fit,nsim=1,time.slices=100)

However, the tergm-related output that would normally be attached to the network (toggles, etc) hasbeen removed.

Source

statnet project stergm tutorial.

Examples

require(network)require(networkDynamic)data(msm.sim)# show the network, aggregating 5 timestepsplot(network.extract(msm.sim,onset=0,terminus=5),

vertex.col=msm.sim%v%"race",vertex.cex=0.5,edge.col="gray")

# this could take 10 minutes, so don't run in example checking## Not run:# use ndtv to render a movie of the momentary view of the network

ndtvAnimationWidget 17

render.animation(msm.sim,vertex.col=msm.sim%v%'race',vertex.cex=.5)ani.replay()saveVideo(ani.replay(),video.name="msm.simMomentary.mp4", other.opts="-b 5000k",clean=TRUE)

# another version, this time with more temporal aggregationmsm.sim <- compute.animation(msm.sim,slice.par=list(start=0,

end=10,interval=1,aggregate.dur=3,rule='latest'))

# also more render more inbetween framesrender.animation(msm.sim,render.par = list(tween.frames = 15,show.times=TRUE),

vertex.col=msm.net%v%'race',vertex.cex=.5)saveVideo(ani.replay(),video.name="msm.sim3Aggregated.mp4", other.opts="-b 5000k",clean=TRUE)

## End(Not run)

ndtvAnimationWidget htmlwidgets wrapper functions for including ndtv-d3 animations inshinyapps

Description

Wrapper functions to provide ndtv-d3 animation as an htmlwidgetfor use within an RStudio "shiny"web application. These functions are not normally called by R users directly. For example shiny apptemplate code please see the ’server.R’ and ’ui.R’ files at https://github.com/statnet/ndtv/tree/master/htmlWidgetShinyTest

Usage

ndtvAnimationWidget(out, options, width = NULL, height = NULL)

renderNdtvAnimationWidget(expr, env = parent.frame(), quoted = FALSE)

ndtvAnimationWidgetOutput(outputId, width = "100%", height = "500px")

Arguments

out the data structure describing the network animation. produced internally byrender.d3movie

options usually the 'd3.options' from render.d3movie

width Display width for the widget. Must be a valid CSS unit (like "100%", "400px","auto") or a number, which will be coerced to a string and have "px" appended.

height Display eight for the widget. Must be a valid CSS unit (like "100%", "400px","auto") or a number, which will be coerced to a string and have "px" appended.

outputId See htmlwidgets-shiny

18 network.layout.animate

expr An expression that does any necessary network processing and generates anHTML widget (usually via render.d3move). See htmlwidgets-shiny

env The environment in which to evaluate expr. See htmlwidgets-shiny

quoted Is expr a quoted expression (with quote())? This is useful if you want to save anexpression in a variable. See htmlwidgets-shiny

Details

The ndtv-d3 interactive HTML5 network animation can normally be produced via render.d3movie(...,output.mode='htmlWidget').These functions are wrappers to make it possible to embed the animations as part of a ’Shiny’(http://shiny.rstudio.com/) web application.

renderNdtvAnimationWidget should be used as the wraper for the render.d3movie call withinthe app’s server.R file and ndtvAnimationWidgetOutput is the corresponding ui component toinclude in the ui.R file. See htmlwidgets-shiny

ndtvAnimationWidget initializes the widget, usually called automatically inside render.d3moviewhen output.mode='htmlWidget'.

Author(s)

skyebend@uw.edu

See Also

htmlwidgets-package, render.d3movie

network.layout.animate

Sequentially-stable network layout algorithms suitable for generatingnetwork animations.

Description

The network.layout.animate.* layouts are often adaptations or wrappers for existing static lay-out algorithms with some appropriate presets. They all accept the coordinates of the ‘previous’layout as an argument so that they can try to construct a suitably smooth sequence of node posi-tions. Usually these layouts are not called directly and instead selected by specifying the appropriateanimation.mode argument to compute.animation

Usage

network.layout.animate.kamadakawai(net, dist.mat = NULL, default.dist = NULL,seed.coords = NULL, layout.par = list(),verbose=FALSE)

network.layout.animate.MDSJ(net, dist.mat = NULL, default.dist = NULL,seed.coords = NULL, layout.par=list(max_iter=50, dimensions =2),verbose=TRUE)

network.layout.animate 19

network.layout.animate.useAttribute(net, dist.mat = NULL, default.dist = NULL,seed.coords = NULL, layout.par = list(x = "x", y = "y"), verbose = TRUE)

network.layout.animate.Graphviz(net, dist.mat = NULL, default.dist = NULL,seed.coords = NULL, layout.par = list(), verbose = TRUE)

Arguments

net The network (or temporal sub-network) that will be used to determine vertexpositions.

dist.mat A (usually optional) matrix of distances between vertices that should be usedto define node positions. This is important to provide if network edge weightsneed special handling - for example to be fliped from similarities to distances,symmetrized, etc.

default.dist A default distance value which a layout may use to fill in for undefined dyads tospace out isolates and disconnected components.

seed.coords A two-column by n-vertex matrix of starting coordinates for the layout to use,usually the coordinates of the previous layout.

layout.par A list of named layout parameters specific to the algorithm.

verbose Print more information about the layout process

Details

These layouts are generally called by compute.animation on a sequence of extracted networks,with each layout fed the output of the previous layout

Usually if the dist.mat is not included, one will be calculated using the layout.distance func-tion which will compute the geodesic path length distance between nodes after symmetrizing thenetwork and replacing Inf values with either sqrt(network.size) or the passed in default.dist

KamadaKawai

The KamadaKawai option provides resonably good dynamically stable layouts. It computes a sym-metric geodesic distance matrix from the input network (replacing infinite values with default.dist,and seeds the initial coordinates for each slice with the results of the previous slice in an attempt tofind solutions that are as close as possible to the previous positions. However, it performs poorly onlarge networks and is not as stable as MDSJ. See network.layout.kamadakawai for more detailsabout the implementation and parameters

MDSJ

The MDSJ layout uses the MDSJ Java library written by Christian Pich, Algorithmics Group, De-partment of Computer & Information Science, University of Konstanz, Germany http://algo.uni-konstanz.de/software/mdsj/(original url), 2009. The library does Multidimensional Scaling (MDS) of the distance matrix usingSMACOF optimization. Because MDSJ is released under a creative commons by-nc-sa license it isnot distributed with the ndtv package, but an installer is included.

When the MDSJ layout is called it checks for working Java installation, and then checks if MDSJis installed. If not, it prompts the user and (optionally) downloads and installs MDSJ. If MDSJ isnot installed, it falls back to calling the KamadaKawai layout instead.

20 network.layout.animate

MDSJ is quite fast for larger networks, but relatively less efficient for smaller ones because ofthe overhead of system calls and Java start up for each layout. The verbose option prints moreinformation on the Java process. The max_iter parameter sets the maximum of minimization stepsthe algorithm can try. In cases where it seems like the layout has not completely finished, thiscan be set higher. The dimensions argument sets the number of dimensions the layout should beperformed in and indirectly the number of columns expected and produced for coordinate matrices.

useAttribute

The useAttribute layout makes it possible to define vertex positions using a static or dynamic ver-tex attributes to provide the x and y coordinates for each time step. The names of the attributes to beused are passed in via the layout.par argument. For example layout.par = list(x = "myX", y = "myY")The attribute must have values defined for each time point that the network plotted.

Graphviz

The Graphviz layout is a wrapper for the Graphviz http://www.graphviz.org software library. Ifthe library is installed on your system (see install.graphviz), it provides a number of additionalhigh-quality layouts. When layout is called it checks for a working Graphviz installation (fallingback to KamadaKawai if Graphviz cannot be found) and writes the network to a temp file usingexport.dot. Then the appropriate Graphviz layout engine (default is neato) is executed via asystem call, and the coordinates of the vertices are parsed from the output.

Currently, the arguments to layout.par can be used to specify the Graphviz layout engine to use(i.e. gv.engine='neato' for stress-minimized, gv.engine='dot' for hierarchal, gv.engine='fdp'for force-directed, etc) and additional command-line control parameters can be passed in via gv.args.For example, to use the ’dot’ layout, but change layout rank direction to Left-Right: layout.par=list(gv.engine='dot',gv.args='-Grankdir=LR').See https://graphviz.gitlab.io/_pages/doc/info/command.html. Note that Graphviz’s graphicrendering parameters are not used to control network plot rendering (but they may impact layoutpositions).

It is also possible to pass edge attributes of the network directly through to the Graphviz .dot file bypassing in the names of the attributes using gv.edge.attrs argument to layout.par. For example,layout.par=list(gv.edge.attrs='len') will write the value of the edge attribute ’len’ to a gvattribute ’len’, which would control the edge lengths when using neato or fdp https://graphviz.gitlab.io/_pages/doc/info/attrs.html#d:len.

The Graphviz layout normally ignores the values in dist.mat, but for compatibility with otherlayouts, it is possible to use the values in dist.mat to influence Graphviz’s edge length by settinglayout.par gv.len.mode='ndtv.distance.matrix' instead of its default 'gv.edge.len'. Thiswrites out all of the possible edges to the file and will overide any other edge attributes.

Value

A two-column by n-vertex matrix of coordinates.

Note

The MDSJ algorithm can only be used for non-comercial projects as it is available under the termsand conditions of the Creative Commons License "by-nc-sa" 3.0. http://creativecommons.org/licenses/by-nc-sa/3.0/

proximity.timeline 21

Author(s)

Skye Bender-deMoll

References

Algorithmics Group. MDSJ: Java Library for Multidimensional Scaling (Version 0.2). Available athttp://algo.uni-konstanz.de/software/mdsj/. University of Konstanz, 2009.

Kamada, T. and Kawai, S. (1989). An Algorithm for Drawing General Undirected Graphs. Infor-mation Processing Letters, 31(1):7-15.

John Ellson et.al (2001) "Graphviz – open source graph drawing tools" Lecture Notes in ComputerScience. Springer-Verlag. p483-484 http://www.graphviz.org

See Also

See Also network.layout.kamadakawai,layout.distance,compute.animation

proximity.timeline Plot a chart of a networkDynamic object in which vertices trace outpaths in time, positioned vertically so that their proximity correspondsto their relative geodesic distance at the sampled time points.

Description

This a DRAFT version of the function, parameters are likely to change. Creates a ’phase plot’ chartof vertex geodesic distance proximities overtime time, with the ability to size and color the linescorresponding to each vertex with arguments similar to plot.network

Usage

proximity.timeline(nd, start = NULL, end = NULL, time.increment = NULL,onsets = NULL, termini = NULL, rule='earliest', default.dist = NULL,vertex.col = "#55555555", label = network.vertex.names(nd),labels.at = NULL, label.cex = 1,vertex.cex = 2, splines = -0.2, render.edges=FALSE, grid=!render.edges,edge.col='#00000055', edge.lwd=4,mode=c('isoMDS','sammon','cmdscale','gvNeato','MDSJ'),coords=NULL,draw.inactive=NULL,spline.style=c('default','inactive.ghost','inactive.gaps',

'inactive.ignore','color.attribute'),chain.direction=c('forward','reverse'),verbose=TRUE, ...)

22 proximity.timeline

Arguments

nd a networkDynamic object to be plotted.

start optional numeric value giving the time to start the network sampling to be passedto link{get.networks}

end optional numeric value giving the time to end the network sampling to be passedto link{get.networks}

time.increment optional numeric value to increment network sampling to be passed to link{get.networks}

onsets optional numeric vector of sampling onset time points to be passed to link{get.networks}

termini optional numeric vector of sampling terminus time points to be passed to link{get.networks}

rule attribute aggregation rule (default 'earliest') to be passed to network.collapse

default.dist numeric default distance parameter to space apart isolates and disconnectedcomponents. Usually defaults to square root of network size (see layout.distance)

vertex.col either a character color value, a vector of values of length equal to the size ofthe network, or the name of a vertex attribute containing color values to be usedto color each of the vertices splines. Note that partially transparent colors workmuch better than opaque colors, as it is easier to see when lines overlap. Whenused with spline.style='color.attribute', vertex.col can be a functionwith a special limited set of arguments (see Details of render.animation)which will be evaluated at the onset of each segment.

labels.at numeric value or vector of values specifying the time(s) at which vertex labelsshould be plotted on the splines. If NULL (default), labels will not be drawn.

label character vector of labels for vertices or name of vertex attribute to be expanded.Default is network.vertex.names. Labels only drawn if labels.at argument hasa value.

label.cex numeric character expansion factor for vertex labels

vertex.cex either a numeric value, a vector of values of length equal to the size of thenetwork, or the name of a vertex attribute containing numeric values to be usedto scale the width of the lines (lwd) for each vertex.

splines numeric. value controls how tightly the splines meet their control points. Avalue of 0 draws straight lines and sharp corners, values less than zero cause thespline to pass through the control point, values greater than zero will approxi-mate the point. See the shape argument of xspline.

render.edges logical (default FALSE). Should overlapping virtical lines corresponding to theedges be drawn between the the splines corresponding to the vertices at the timepoints of edge onsets?

edge.col color value or edge attribute name to be used for the edge lines if render.edges=TRUE

edge.lwd numeric line width value or edge attribute name to be used for the width of theedge lines if render.edges=TRUE

grid logical. if TRUE, vertical lines in the background color will be drawn at thebeginning of each time slice to make it easier to determine where on the splinesthe positions are actually set. Usually this is not used with render.edges

mode name of MDS algorithm to be used. Currently one of isoMDS,sammon,cmdscale

proximity.timeline 23

coords optional numeric matrix of pre-computed coordinates to be used instead of thealgorithm in mode. The number of matrix rows must be equal to the networksize and columns equal to number of time bins implied by other arguments)

draw.inactive DEPRECATED. see spline.style

spline.style options to control how vertices with inactive spells or changing attribute valuesshould be drawn:

• 'inactive.ignore' ignores activity spells and draws an unbroken splinefor each vertex (fastest).

• 'inactive.gaps' leaves gaps in the splines when vertices are inactive• 'inactive.ghost' draws faint gray dotted lines under the spline so they

appear in the gaps• 'default' does 'inactive.ignore' if there are no gaps in encountered,

otherwise 'inactive.ghost'

• 'color.attribute' uses the activity spells of the vertex color TEA (in-dicated by the vertex.col argument) to break the splines in to color seg-ments – ignoring the the vertices activity spells

chain.direction

value of 'forward' means that the slice layouts should be computed in temporalorder, with each layout initialized with the coordinates from the previous. Avalue of 'reverse' causes layouts to be computed in reverse temporal order(for some layouts, this will cause less spline crossing as vertices will tend to becloser to their final state).

verbose logical, default is TRUE, in which case status messages about the computationsare printed to the console, at some speed cost

... arguments to be passed to network.collapse (via get.networks) to controlhow the network should be aggregated during slicing

Details

The passed network dynamic object is sliced up into a series of networks. It loops over the networks,converting each to a distance matrix based on geodesic path distance with layout.distance. Thedistances are fed into an MDS algorithm (specified by mode) that lays them out in one dimension:essentially trying to position them along a vertical line. The sequence of 1D layouts are arrangedalong a timeline, and a spline is drawn for each vertex connecting its positions at each time point.The idea is that closely-linked clusters form bands of lines that move together through the plot.

Currently,

• mode='sammon' tends to produce much equally spaced lines, making it easier to follow indi-vidual vertices, but harder to see clusters

• mode='isoMDS' does a better job with clusters, but in some layouts converges too soon andjust produces straight lines,

• mode='cmdscale' does a great job with clusters, but is highly unstable (coordinates willreshuffle dramatically on nearly identical networks).

• mode='gvNeato' tries to do a 1D Graphviz neato layout (experimental) network.layout.animate.Graphviz.

• mode='MDSJ' tries a 1D network.layout.animate.MDSJ layout.

24 proximity.timeline

For most of the layouts it is necessary to manually adjust the default dist parameter to find a valuethat sufficently groups together linked clusters and spaces out isolates.

Note for RStudio users: the spline rendering seems to be much slower on RStudio’s graphics devicethan on other graphics devices such as x11().

Value

Produces a plot with horizontal splines corresponding the vertices of the network and vertical prox-imities approximately proportional to geodesic distance. Invisibly returns a numeric matrix of co-ordinates corresponding to computed positions of each vertex at each time bin. This can be passedin via the coords argument.

Note

This is still very much a work in progress, the 1D optimization are not very stable, especially forcmdscale

Author(s)

skyebend@uw.edu

References

Some inspirational examples here: http://skyeome.net/wordpress/?p=604

See Also

See also timeline for plotting spells of vertices and edges without proximity positioning.

Examples

# use the classroom interaction datasetdata(McFarland_cls33_10_16_96)

# divide the first 20 minutes of time into# overlapping 2.5 minute bins# and make the lines for the instructors much largerproximity.timeline(cls33_10_16_96,

onsets=seq(0,20,0.5),termini=seq(2.5,22.5,0.5),vertex.cex=(cls33_10_16_96%v%'type'=='instructor')*4+1,labels.at=16)

# load the infection sim datasetdata(toy_epi_sim)# render a timeline with vertices colored by infection status# show only the first 5 timestepsproximity.timeline(toy_epi_sim,vertex.col = 'ndtvcol',

spline.style='color.attribute',mode='sammon',default.dist=20,chain.direction='reverse',start=1,end=5)

render.animation 25

render.animation Render animations of networkDynamic objects as movies in variousformats

Description

Takes a network object which describes a sequence of changes in properties of a network andgraphically renders it out as a sequence of plots to create an animation. Normally the coordinatesdetermining the vertex positions at specified time points should have been calculated and storedin the network object, along with the slice.par list of parameters describing when and how thenetwork should be divided up in time. If the coordinate data is not found, compute.animation willbe called with default arguments.

Appropriate ‘static’ networks for each time region will be generated by network.collapse. Therendering of each frame is drawn by plot.network and most arguments are supported and arepassed through to control the details of rendering. The rendered images are stored using theanimation package and can be played in the plot window (ani.replay) or saved to a movie filewith saveVideo.

Usage

render.animation(net, render.par = list(tween.frames = 10, show.time = TRUE,show.stats = NULL, extraPlotCmds=NULL, initial.coords=0),plot.par = list(bg='white'), ani.options = list(interval=0.1),render.cache = c('plot.list','none'), verbose=TRUE, ...)

Arguments

net The networkDynamic object to be rendered, usually containing pre-computedvertex positions as dynamic attributes.

render.par Named list of parameters to specify the behavior of the animation.

• tween.frames the number of interpolated frames to generate between eachcalculated network layout (default 10).

• show.time If TRUE, labels the plot with onset and terminus time for eachslice.

• show.stats NULL, or a string containing a formula to be passed to sum-mary.stergm to display the network statistics for the current slice on theplot. e.g. "~edges+gwesp(0,fixed=TRUE)"

• extraPlotCmds NULL, or additional plot commands to draw on each frameof the animation.

• initial.coords default initial coords to be assigned to vertices. Can be atwo-column numeric coordinate matrix, or a numeric values to be formedinto a matrix.

plot.par list of ‘high-level’ plotting control arguments to be passed to par. e.g. bg forbackground color, margins, fonts, etc.

26 render.animation

ani.options list of control arguments for the animation library. For example. intervalcontrols the delay between frames in playback, ani.dev and ani.type can beused to set non-default graphics devices for rendering (i.e 'jpeg' instead of'png'). See ani.options

render.cache the default value of 'plot.list' causes each frame of the animation to becached in an internal list by the ani.record function of the animation library.This is very useful for testing and replaying animations in R’s plot window, butcan be very slow (or cause out-of-memory errors) for large animations. If thevalue is set to 'none', the plot will not be recorded (but can be saved to disk viasaveVideo, see examples below) and cannot be replayed via the ani.replay()function.

verbose If TRUE, update text will be printed to the console to report the status of therendering process.

... Other parameters to control network rendering. See plot.network.default.Most parameters can be set to TEA attribute names, or specified as a function tobe evaluated at each timestep

Details

Most of the network plotting arguments are passed directly to plot.network.default. All of theplot.network arguments pased in via ... can be specified as a TEA or special type of functionto be evaluated at each time step. For example, if there was a dynamic vertex attribute named’wealth’, it could be mapped to vertex size by providing the TEA name vertex.cex='wealth'.If the wealth value needed transformation to be an appropriate vertex size, it can be specified as afunction: vertex.cex=function(slice){slice%v%'wealth'*5-3}

The arguments of plot control functions must draw from a specific set of named arguments whichwill be substituted in and evaluated at each time point before plotting. The set of valid argumentnames is:

• net is the original (uncollapsed) network

• slice is the network collapsed with the appropriate onset and terminus

• s is the slice number

• onset is the onset (start time) of the slice to be rendered

• terminus is the terminus (end time) of the slice to be rendered

A few of the plot parameters have defaults that are different from the ones given by plot.network:

• jitter defaults to FALSE to prevent unwanted bouncing

• xlim and ylim default to the ranges of the entire set of network coordinates. Although theycan be set to function values for interesting effects...

• xlab defaults to a function to display the time range: function(onset,terminus){paste("t=",onset,"-",terminus,sep='')}.It also will be overidden if show.stats is set.

If no slice.par network attribute is found to define the time range to render, it will make one upusing the smallest and largest non-Inf time values, unit-length non-overlapping time steps and anaggregation rule of ’latest’.

render.animation 27

If no dynamic coordinate information has been stored, compute.animation will be called withdefault values to try to do the layout before giving up.

Additional plot commands passed in via the extraPlotCmds argument will be passed to eval()after each frame is rendered and can be used to add extra annotations to the plot.

One some installations, the default output from saveVideo() (really ffmpeg) produces videos in aslightly non-standard .mp4 format that won’t play in Windows Media Player or QuickTime. Usershave reported that adding the argument other.opts='-pix_fmt yuv420p" to saveVideo corrects theproblem. Recent versions of the animation library will include this argument by default.

To avoid performance issues with the RStudio graphics device, RStudio users will see a message thatndtv is attempting to open another type of plot window. It will try to guess a platform-appropriatedevice type, but specific device can be pre-specified by the user by setting the R_DEFAULT_DEVICEenvironment variable

Value

A sequence of plots will be generated on the active plot device. If render.cache='plot.list'the recorded plots are stored as a list in .ani.env$.images.

Note

A few of the network drawing arguments have slightly different interpretations than their plot.networkequivalents:

• xlab will be used to display time and network statistics if those render.par parameters areset

• xlim and ylim will be set using max and min observed coordinate values so that all networkslices will appear on the plot

• label if not set explicitly, will default to the vertex names for the appropriate slice network.

If the background color is transparent and not explicitly set, it will be reset to white to preventunintentional behavior when exporting movies via ffmpeg.

Author(s)

Skye Bender-deMoll, and the statnet team.

References

Skye Bender-deMoll and McFarland, Daniel A. (2006) The Art and Science of Dynamic NetworkVisualization. Journal of Social Structure. Volume 7, Number 2 http://www.cmu.edu/joss/content/articles/volume7/deMollMcFarland/

See Also

compute.animation for generating the movie coordinates, ani.replay, plot.network and thepackage vignette vignette('ndtv'). Also render.d3movie for displaying movies as interactiveHTML5 animations in a web browser

28 render.animation

Examples

require(ndtv)# trivial example

triangle <- network.initialize(3) # create a toy networkadd.edge(triangle,1,2)# add an edge between vertices 1 and 2add.edge(triangle,2,3)# add a more edgesactivate.edges(triangle,at=1) # turn on all edges at time 1 onlyactivate.edges(triangle,onset=2, terminus=3,e=get.edgeIDs(triangle,v=1,alter=2))add.edges.active(triangle,onset=4, length=2,tail=3,head=1)render.animation(triangle)ani.replay()

# an example with changing TEA attributeswheel <- network.initialize(10) # create a toy networkadd.edges.active(wheel,tail=1:9,head=c(2:9,1),onset=1:9, terminus=11)add.edges.active(wheel,tail=10,head=c(1:9),onset=10, terminus=12)# set a dynamic value for edge widthsactivate.edge.attribute(wheel,'width',1,onset=0,terminus=3)activate.edge.attribute(wheel,'width',5,onset=3,terminus=7)activate.edge.attribute(wheel,'width',10,onset=3,terminus=Inf)# set a value for vertex sizesactivate.vertex.attribute(wheel,'mySize',1, onset=-Inf,terminus=Inf)activate.vertex.attribute(wheel,'mySize',3, onset=5,terminus=10,v=4:8)# set values for vertex colorsactivate.vertex.attribute(wheel,'color','gray',onset=-Inf,terminus=Inf)activate.vertex.attribute(wheel,'color','red',onset=5,terminus=6,v=4)activate.vertex.attribute(wheel,'color','green',onset=6,terminus=7,v=5)activate.vertex.attribute(wheel,'color','blue',onset=7,terminus=8,v=6)activate.vertex.attribute(wheel,'color','pink',onset=8,terminus=9,v=7)# render it allrender.animation(wheel,edge.lwd='width',vertex.cex='mySize',vertex.col='color')

# an example with functional attributesset.network.attribute(wheel,'slice.par',

list(start=1,end=10,interval=1, aggregate.dur=1,rule='latest'))# render vertex size as betweenessrender.animation(wheel,vertex.cex=function(slice){(betweenness(slice)+1)/5})

# render it directly to a movie file without caching the plots (faster)## Not run:saveVideo( render.animation(wheel,edge.lwd='width',vertex.cex='mySize',vertex.col='color',

render.cache='none') )

## End(Not run)

render.d3movie 29

# simulation based example# disabled to save time when testing## Not run:require(tergm)# load in example data, results of a basic stergm simdata(stergm.sim.1)

# (optional) pre-compute coordinates for set time range# (optional) limit time range to a few steps to speek exampleslice.par=list(start=0,end=10,interval=1, aggregate.dur=1,rule='latest')compute.animation(stergm.sim.1,slice.par=slice.par)

# define the number of inbetween frames and a formula for stats to displayrender.par<-list(tween.frames=5,show.time=TRUE,

show.stats="~edges+gwesp(0,fixed=TRUE)")

# render the movie, with labels, smaller vertices, etcrender.animation(stergm.sim.1,render.par=render.par,

edge.col="darkgray",displaylabels=TRUE,label.cex=.6,label.col="blue")

# preview the movie in the plot windowani.replay()

# save the movie as mp4 compressed video (if FFMPEG installed)saveVideo(ani.replay(),video.name="stergm.sim.1.mp4",

other.opts="-b 5000k",clean=TRUE)

## End(Not run)

render.d3movie Render out a web-based animation of a networkDynamic object usingndtv-d3 player app

Description

Exports a self-contained HTML file including an SVG animation of the networkDynamic objectand displays it in a web browser. See render.animation for details of animation construction.

Usage

render.d3movie(net, filename=tempfile(fileext = '.html'),render.par=list(tween.frames=10,

show.time=TRUE,show.stats=NULL,extraPlotCmds=NULL,initial.coords=0),

30 render.d3movie

plot.par=list(bg='white'),d3.options,output.mode=c('HTML','JSON','inline','htmlWidget'),script.type=c('embedded','remoteSrc'),launchBrowser=TRUE,verbose=TRUE,...)

Arguments

net The network (usally networkDynamic) object to be rendered, usually containingpre-computed vertex positions as dynamic attributes cached by compute.animation

filename The file name of the HTML or JSON file to be generated. Must end the properfile extension (’.json’ for JSON ’.html’ for HTML) for correct browser display.

render.par Named list of parameters to specify the behavior of the animation.

• tween.frames the number of interpolated frames to generate between eachcalculated network layout (default 10).

• show.time If TRUE, labels the plot with onset and terminus time for eachslice.

• show.stats NULL, or a string containing a formula to be passed to sum-mary.stergm to display the network statistics for the current slice on theplot. e.g. "~edges+gwesp(0,fixed=TRUE)"

• extraPlotCmds NULL, or additional plot commands to draw on each frameof the animation.

• initial.coords default initial coords to be assigned to vertices. Can be atwo-column numeric coordinate matrix, or a numeric values to be formedinto a matrix.

plot.par list of ‘high-level’ plotting control arguments to be passed to par. e.g. bg mainfor background color, margins, fonts, etc. MANY OF THESE ARE NOT YETSUPPORTED BY THE NDTV-D3 PLAYER. See list in Details below.

d3.options list of options to configure ndtv-d3 player app. list( animationDuration=800, scrubDuration=0, enterExitAnimationFactor=0, nodeSizeFactor=0.01, playControls=TRUE, animateOnLoad=FALSE, slider=TRUE, debugFrameInfo=FALSE, debugDurationControl=FALSE, )See Details below for explanations

output.mode character string, one of 'HTML','JSON' or 'inline'. The first exports an HTMLfile with embedded javascript player app including the JSON data structure de-scribing the animation. The second just exports the JSON data structure (forloading into an existing page). The 3rd renders the HTML inside an iframetag and supresses all other output in an attempt to make it embedable in rmark-down documents. The 4th generates a htmlWidget suitable for displaying in anRStudio plot window or Shiny app.

script.type if script.type='embedded', the scripts will be embedded directly in the out-put html page. This option is the most portable, but will require large file sizes.If script.type='remoteSrc', only the links to the scripts will be included, sothe page will require an active internet connection to play the animation.

launchBrowser if TRUE, after exporting the file R will attempt to open it in a browser

verbose If TRUE, update text will be printed to the console to report the status of therendering process.

render.d3movie 31

... Other parameters to control network rendering. See plot.network.default.Most parameters can be set to TEA attribute names, or specified as a functionto be evaluated at each timestep. NOT ALL PLOT PARAMS ARE IMPLE-MENTED YET

Details

Animations are generated using a process nearly identical to render.animation. However, insteadof using R’s plotting functions and the animation library, the relevant information is cached andwritten into a JSON-formatted file, embedded into a web page along with ndtv-d3 player, anddisplayed in a web browser as an interactive HTML5 SVG animation.

The ndtv-d3 player app is not a fully-featured R plot device. It only attempts to emulate theelements of plot normally used by plot.network and it understands the graphic elements and asomewhat higher level so that it will be able to handle interaction with edge and vertex objects.

However, ndtv-d3 includes several nice features to support exploring the network:

• controller buttons for playing, pausing and stepping through the animation

• time slider for jumping and ’scrubing’ to parts of the movie

• pan and zoom into the network using the mouse-wheel

• click on vertices and edges to reveal their ids or abitrary text attached using the vertex.tooltipand edge.tooltip properties

• double-click on a vertex to highlight all of the connected edges and vertices

If passed a static network, by default only a single slice will be rendered and the time sliderand controllers will be disabled. For consistency with plot.network the static mode also sup-ports passing in a matrix of coordinates via coord argument which will prevent the default call tocompute.animation.

Another advantage of ndtv-d3 is that it does not require installing system libraries such as ffmpegto render out the movie.

The coordinates for vertex postitions are read from the animation.x and animation.y TEA at-tributes, normally created using compute.animation

The list of currently supported plot,plot.network and par elements is

• xlab: label caption below the render, on the xaxis

• main: main headline above the render

• displaylabels: should vertex labels be displayed?

• usearrows: should arrows be drawn on edges?

• bg: background color (must be html compatible? need to check this)

• vertex.cex: vertex expansion scale factor

• label: labels for vertices (defaults to vertex.names)

• label.col: color of vertex labels

• label.cex: vertex label expansion scale factor

• vertex.col: vertex fill colors

• vertex.sides: number of sides for vertex polygon (shape)

32 render.d3movie

• vertex.rot: rotation for vertex polygon

• vertex.border: color of vertex border stroke

• vertex.lwd: width of vertex border stroke

• edge.lwd: width of edge stroke

• edge.col: edge stroke color

All of the above properties can be defined as dynamic (TEA) attributes. Noteably, curved edges,edge labels, and label positioning are not yet implemented and will be ignored. The main and xlabparams will not be positioned exactly as they are in plot

There are a few special plot parameters that are only supported by render.d3move:

• vertex.tooltip arbitrary text or html to be displayed when a vertex is clicked (default is thevertex id)

• edge.tooltip arbitrary text or html to be displayed when an edge is clicked (default is theedge id)

• vertex.css.class properties for adding arbitrary class attributes for use in CSS styling ofvertices

• edge.css.class properties for adding arbitrary class attributes for use in CSS styling ofedges

• vertex.label.css.class properties for adding arbitrary class attributes for use in CSSstyling of vertex labels

ndtv-d3 has its own configuration properties passed in via the d3.options argument list, shownbelow with their default properties. Values which are set to NULL or omitted will be set to thendtv-d3 player defaults.

• animationDuration=800 Duration (milliseconds) of each animation step during play or stepactions

• enterExitAnimationFactor=0 Fraction (0-1) of total step animation time that edge en-ter/exit animations should take

• nodeSizeFactor=0.01 Sets default node (vertex) size, as a fraction of viewport size.

• playControls=TRUE Show the player controls (play, pause, step, etc)

• animateOnLoad=FALSE If true, animation will start playing as soon as page loads.

• slider=TRUE Show the time slider control

• margin=list(x=20,y=10) SVG margins - may be overridden when setting fixed aspect ratio

• debugFrameInfo=FALSEShow the slice timing info in corner

• durationControl=TRUEShow a control to change speed of animation under the menu in theupper right corner

Value

If output.mode='HTML', a file will be generated including the necessary javascript and JSON datastructures If output.mode='JSON', a JSON file will be generated including a section describingall of the rendered slices, and a seperate section including the entire networkDynamic object.

render.d3movie 33

If output.mode='inline', HTML code for an iframe elment suitible for embedding in mark-down documents will be printed, all other output supressed. If output.mode='htmlWidget', ahtmlwidgets object will be returned which, will produce appropriate html when ’printed’ or emed-ded in a Shiny app

Note

This is a very preliminary draft implementation. The animations preform poorly in the Linux Fire-fox browser, but are ok in Firefox on other platforms and excellent in the Chrome web browser.

Author(s)

skyebend@uw.edu

References

The github repository for the ndtv-d3 javascript library is at https://github.com/statnet/ndtv-d3/(which is the statnet release fork of https://github.com/michalgm/ndtv-d3/)

See Also

See also the ndtv-d3 vignette http://statnet.csde.washington.edu/workshops/SUNBELT/current/ndtv/ndtv-d3_vignette.html, render.animation, compute.animation.

Examples

# render an interactive SVG animation of short.stergm.sim and open it in a browserdata(short.stergm.sim)render.d3movie(short.stergm.sim)

# render interactive widget in rmarkdown or RStudio plot windowrender.d3movie(short.stergm.sim,output.mode='htmlWidget')

# render a static network as interactive SVG with lots of html tooltip infodata(emon)render.d3movie(emon[[5]],

vertex.tooltip=paste(emon[[5]]%v%'vertex.names',emon[[5]]%v%'Command.Rank.Score',emon[[5]]%v%'Sponsorship',sep="<br>"),

edge.tooltip=paste('Frequency:',emon[[5]]%e%'Frequency'),edge.lwd='Frequency')

## Not run:

# alternate code for embeding in rmarkdown```{r,results='asis'}render.d3movie(short.stergm.sim,output.mode='inline')```

34 stergm.sim.1

## End(Not run)

stergm.sim.1 Very Very Basic stergm simulation output

Description

Simulation from a crude stergm model based on the flobusiness network. Mostly good for testingmovies ’cause it is small (16 vertices) and fast. The stergm.sim.1 network is 100 simulation stepsin duration. The short.stergm.sim network is an extract of the first 25 steps of stergm.sim.1 –its shorter duration makes it more suitable for quickly testing animation techniques.

Usage

data(stergm.sim.1)data(short.stergm.sim)

Format

A networkDynamic object containing the output of the network simulations

Details

The model used to generate the sim was:

require(ergm)data("florentine")theta.diss <- log(9)# fit the modelstergm.fit.1 <- stergm(flobusiness,

formation= ~edges+gwesp(0,fixed=T),dissolution = ~offset(edges),

targets="formation",offset.coef.diss = theta.diss,estimate = "EGMME" )# simulate from the modelstergm.sim.1 <- simulate(stergm.fit.1,

nsim=1, time.slices = 100)

However, the ergm-related output that would normally be attached to the network (toggles, etc) hasbeen removed.

Source

See tergm package tutorials.

timeline 35

Examples

data(stergm.sim.1)range(get.change.times(stergm.sim.1))data(short.stergm.sim)range(get.change.times(short.stergm.sim))

timeline Plot a timeline for the edge and vertex spells of a network

Description

Produces a ‘phase plot’ or timeline showing the durations of the activity spells in a networkDynamicobject. Spells are traced out horizontally, with all the activity for each element (vertex or edge) in asingle row.

Usage

timeline(x, v = seq_len(network.size(x)), e = seq_along(x$mel),plot.vertex.spells = TRUE, plot.edge.spells = TRUE,slice.par = NULL,displaylabels = TRUE, e.label=TRUE, e.label.col='purple',edge.lwd=1,v.label, v.label.col='blue',vertex.cex=1, cex, adj=0,edge.col = rgb(0.5, 0.2, 0.2, 0.5),vertex.col = rgb(0.2, 0.2, 0.5, 0.5),xlab, ylab, xlim, ylim, ...)

Arguments

x a networkDynamic object that will have its spells plotted.

v numeric vector of vertex ids to include

e numeric vector of edge ids to includeplot.vertex.spells

logical, should vertex spells be plotted?plot.edge.spells

logical, should edge spells be plotted?

slice.par (optional) ‘slice.par’ list giving network binning parameters. If included, rectan-gles corresponding to each bin will be plotted over the spells to indicate whichspell will land in bins. The bins will be drawn with slightly darker left edgemore transparent right edge to evoke the effect of a right-open interval.

displaylabels logical, should labels be drawn for each spell

e.label character vector of edge labels or edge attribute name. Default is edge.id

36 timeline

e.label.col color name or character vector of colors for edge labels (or name of edge at-tribute to provide them)

v.label character vector of vertex labels or vertex attribute name. Default is network.vertex.names

v.label.col color name or character vector of colors for vertex labels (or name of vertexattribute to provide them)

vertex.cex numeric scaling factor, vector of numeric scaling factors or attribute name. Trans-lated width of line (lwd) corresponding to each vertex.

edge.lwd numeric scale factor, numeric vector, or character edge attribute name providinga numeric value for the width of the lines corresponding to each edge. Notethat this does not behave exactly as edge.lwd in plot.network as it does notperform scaling based on attribute values.

cex text size scaling for both vertex and edge labels (see plot.default)

adj text justification parameter (see par) for both vertex and edge labels. Labels arepositioned relative to onset of spell.

edge.col color to be used to draw lines for edge spells, or vector of color names corre-sponding to edges, or name of edge attribute.

vertex.col color to be used to draw lines for vertex spells, or vector of color names corre-sponding to vertices, or name of vertex attribute.

xlab x-axis label for plot

ylab y-axis label for plot

xlim two-element numeric vector giving the x-range (time bounds) of the plot toshow. Defaults to (non-infinite) max and min time of network.

ylim two-element numeric vector giving the y-range (effectively the number of en-tities) of plot to show. Defaults to an appropriate mapping of the number ofentities to the available plot size.

... additional arguments to be passed to plot subroutines. See plot.default,lines,text.

Details

When the v argument is included, edges involving vertices not in v are excluded (but the reverse isnot true for the e argument). If xlim range is provided and the spells corresponding to a vertex oran edge lie entirely outside its bounds they will not be shown.

Many of the arguments correspond to arguments in plot.network but are translated to the timelineplot context. For example, vertex.cex actually controls the lwd (line width) of the lines cor-responding to vertex spells. The arguments are expanded using plotArgs.network so that theyshould give the expansion behavior and attribute look up as plot.network

Additional plotting arguments can be passed in to modify drawing. For example, lty for line style.Vertices and edges that are never active are not included on the plot.

Value

A plot is produced.

timePrism 37

Note

not fully implemented, would be nice to be able to pass network attribute names for properties..

Author(s)

skyebend@uw.edu

See Also

See also plot.network.

Examples

data(stergm.sim.1)timeline(stergm.sim.1)

# color vertices by priorates, don't show edgestimeline(stergm.sim.1,vertex.col='priorates',plot.edge.spells=FALSE)

# show only relationships among a few verticestimeline(stergm.sim.1,v=1:8)

# zoom in on a region of timetimeline(stergm.sim.1,xlim=c(20,40))

# label vertices with numbers# and label edges by the tail and head vertices they linktimeline(stergm.sim.1,xlim=c(0,5),v.label=1:network.size(stergm.sim.1),

e.label=sapply(stergm.sim.1$mel,function(e){paste(e$inl,e$outl,sep='->')}) )

# show only edge spells, hi-lite edge id 20set.edge.attribute(stergm.sim.1,'my_color','gray')set.edge.attribute(stergm.sim.1,'my_color','red',e=20)timeline(stergm.sim.1,edge.col='my_color',plot.vertex.spells=FALSE)

# show binning over the edgestimeline(stergm.sim.1,slice.par=list(start=0,

end=100,interval=10, aggregate.dur=5,rule='latest'),

plot.vertex.spells=FALSE)

timePrism Plot a networkDynamic object as sequence of snapshots in a pseudo-3D space-time prism

38 timePrism

Description

Plots an image using scatterplot3d to render multiple network layout ’slices’ in a ’3D’ ortho-graphic projection in which one axis is time. The coordinates for the networks are assumed to havebeen generated by compute.animation

Usage

timePrism(nd, at,spline.v = NULL,spline.col = "#55555555",spline.lwd = 1,box = TRUE,axis = TRUE,planes = FALSE,plane.col = "#FFFFFF99",scale.y = 1,angle = 20,orientation = c("x", "y", "z"),...)

Arguments

nd a networkDynamic object to be plotted

at a numeric vector of times at which the network should be sampled and plotted.

spline.v optional integer vector of vertex ids to highlight with a spline linking the ver-tices” positions at multiple time points

spline.col vector of colors corresponding to spline.v

spline.lwd numeric line width for vector splines

box a logical value indicating whether a box should be drawn around the plot toindicate its bounds

axis a logical value indicating whether the x, y, and z axis should be drawn.

planes a logical value indicating whether a ’plane’ should be drawn to indicate theboundaries of each individual network plot

plane.col a color value to be used to color the planes (usually partially transparent)

scale.y numeric value giving the relative scale of y axis related to x- and z axis. (maydistort network vertex shapes)

angle numeric angle (degrees) between x and y axis (Attention: result depends onscaling).

orientation three-element charter vector the permutation of which determines the mappingand orientation of the plot axis relative to the figure. i.e. default c('x','y','z')will place ’z’ (the time dimension) ’vertically’ up the page, c('z','x','y')will make the time dimension horizontal, etc.

... additional parameters to plot.network

timePrism 39

Details

Implements a common conceptualization of dynamic networks a series of ’layers’ or ’slices’ intime. Mostly useful for illustrative purposes as this plot type tends to get really crowded if morethan a few network time points are shown, or vertices highlighted.

Value

invisibly returns the result of the scatterplot3d command, which contains useful functions as$xyz.convert which can be used to convert xyz coordinates into the plot space for additionalannotation.

Note

Not all of the useful argument passthroughs to scatterplot3d and xspline have been implementedyet. Shapes of vertices and edges can be improperly distorted by coordinate projection.

Author(s)

skyebend@uw.edu

See Also

compute.animation,scatterplot3d, xspline, plot.network. Also filmstrip and proxmity.timelinefor related static views.

Examples

data("short.stergm.sim")compute.animation(short.stergm.sim)timePrism(short.stergm.sim,at=c(1,10,20),

displaylabels=TRUE,label.cex=0.5)

data(toy_epi_sim)timePrism(toy_epi_sim,

orientation=c('z','y','x'),angle=40,spline.v=c(7, 29, 36, 70, 82, 96), # hilite the infectedspline.col='red',spline.lwd=2,box=FALSE,planes=TRUE,vertex.col='ndtvcol')

40 toy_epi_sim

toy_epi_sim Toy Epidemic Simulation Output from the EpiModel package

Description

An example network of a trivial simulated disease process spreading over a simulated dynamiccontact network among 100 individuals for 25 discrete time steps.

Usage

data("toy_epi_sim")

Format

The format is a networkDynamic object with attached attributes for vertex.pid (persistand ids),and dynamic attributes for ndtvcol (color corresponding to infection status) and testatus (infec-tion status of vertices)

Details

The toy_epi_sim network is example output from a basic dynamic network STERGM simulationand trivial "SI" infection simulation generated using the EpiModel package. The model had random("edges only") edge formation and dissolution effects, with rates calculated to lead to mean edgedurations of 10 time units. The infection simulation had an infection probability of 0.8.

The simulation was generated with the following code:

library(EpiModel)

## Network Estimation (using a tergm model)nw <- network.initialize(n = 100, directed = FALSE)formation <- ~ edgestarget.stats <- 50dissolution <- ~ offset(edges)coef.diss <- dissolution_coefs(dissolution, duration = 10)est <- netest(nw,

formation,dissolution,target.stats,coef.diss,verbose = FALSE)

## Epidemic simulationparam <- param.net(inf.prob = 0.8)init <- init.net(i.num = 5)control <- control.net(type = "SI", nsteps = 25, nsims = 1, verbose =

FALSE)sim <- netsim(est, param, init, control)

toy_epi_sim 41

## Use some of EpiModel's default coloring functions to cache colorstoy_epi_sim <- get_network(sim)toy_epi_sim <- color_tea(toy_epi_sim)

References

Samuel Jenness, Steven M. Goodreau and Martina Morris (2015). EpiModel: Mathematical Mod-eling of Infectious Disease. R package version 1.1.4. https://CRAN.R-project.org/package=EpiModel

Statnet EpiModel Tutorial http://www.epimodel.org/

See Also

See also short.stergm.sim for another basic Stergm simulation output, and msm.sim for a largerand more complex simulation without an infection process.

Examples

data(toy_epi_sim)timeline(toy_epi_sim)## Not run:

# set up layout to draw plots under timelinelayout(matrix(c(1,1,1,2,3,4),nrow=2,ncol=3,byrow=TRUE))# plot a proximity.timeline illustrating infection spreadproximity.timeline(toy_epi_sim,vertex.col = 'ndtvcol',

spline.style='color.attribute',mode = 'sammon',default.dist=100,chain.direction='reverse')

# plot 3 static cross-sectional networks# (beginning, middle and end) underneath for comparisonplot(network.collapse(toy_epi_sim,at=1),vertex.col='ndtvcol',

main='toy_epi_sim network at t=1')plot(network.collapse(toy_epi_sim,at=17),vertex.col='ndtvcol',

main='toy_epi_sim network at=17')plot(network.collapse(toy_epi_sim,at=25),vertex.col='ndtvcol',

main='toy_epi_sim network at t=25')layout(1) # reset the layout

# render an animation of the networkrender.animation(toy_epi_sim,vertex.col='ndtvcol',displaylabels=FALSE)ani.replay()

## End(Not run)

42 transmissionTimeline

transmissionTimeline plots network diffusion/transmission tree with generation time vs.clock/model time

Description

Plots view of a network with positions determined by the timing and generation depth (previouslycalculated) in a transmission tree. The horizontal axis is model time, and the vertical axis is thenumber of steps from the root of the tree.

Usage

transmissionTimeline(x, time.attr,label,displaylabels = !missing(label),label.cex = 0.7,label.col = 1,vertex.col = 2,vertex.border = 1,vertex.lwd = 1,vertex.sides = 50,vertex.cex = 1,jitter=FALSE,edge.col = "gray",edge.lty = 1,edge.lwd = 1,xlab = "time",ylab = "generation",...)

Arguments

x an object than can be coerced into a network. The network must be a tree

time.attr name of a vertex attribute containing the transmission/infection/diffusion timefor each vertex

label a vector of vertex labels, if desired; defaults to the vertex labels returned bynetwork.vertex.names. If label has one element and it matches with a vertexattribute name, the value of the attribute will be used. Note that labels may beset but hidden by the displaylabels argument.

displaylabels boolean; should vertex labels be displayed?

label.cex character expansion factor for label text.

label.col color for vertex labels; may be given as a vector or a vertex attribute name, iflabels are to be of different colors.

vertex.col color for vertices; may be given as a vector or a vertex attribute name, if verticesare to be of different colors.

transmissionTimeline 43

vertex.border border color for vertices; may be given as a vector or a vertex attribute name, ifvertex borders are to be of different colors.

vertex.lwd line width of vertex borders; may be given as a vector or a vertex attribute name,if vertex borders are to have different line widths.

vertex.sides number of polygon sides for vertices; may be given as a vector or a vertex at-tribute name, if vertices are to be of different types. NOTE: only values of3,4,and 50 (circle) are used as they are translated to pch plot symbols.

vertex.cex expansion factor for vertices; may be given as a vector or a vertex attribute name,if vertices are to be of different sizes.

jitter if TRUE, noise will be added to the coordinates with jitter to make overlappingvertex positions more noticeable

edge.col color for edges; may be given as a vector, adjacency matrix, or edge attributename, if edges are to be of different colors.

edge.lty line type for edge borders; may be given as a vector, adjacency matrix, or edgeattribute name, if edge borders are to have different line types.

edge.lwd line width scale for edges; May be given as a vector or edge attribute name, ifedges are to have different line widths.

xlab y-axis plot label

ylab x-axis plot label

... additional arguments to plot (and par)

Details

Many (but not all) of the graphical arguments to plot.network can be used and are expanded in thesame way. This does not currently use the plot.network code to draw the network as non-squareplot aspect ratios would cause distortion of the vertices when drawn.

Value

produces a plot, invisibly returns the coordinates of the plot.

Author(s)

skyebend@uw.edu

See Also

plot.network, proximity.timeline

Examples

# an edgelist describing an infection treeel <-cbind(c(16, 13, 13, 10, 13, 16, 10, 13, 1, 10, 8, 1, 4, 4, 2, 2),

1:16)# a vector of infection timesinfectionTimes <- c(583, 494, 634, 40, 712, 701, 224, 719,

674, 0, 749, 621, 453, 665, 709, 575)

44 transmissionTimeline

# make a network object, include the infection timeinfTree<-network(el,vertex.attr = list(infectionTimes),

vertex.attrnames = list('infectionTimes'))

transmissionTimeline(infTree,time.attr='infectionTimes')

Index

∗Topic IOexport.dot, 7export.pajek.net, 9

∗Topic datasetsmsm.sim, 15stergm.sim.1, 34toy_epi_sim, 40

∗Topic packagendtv-package, 2

ani.options, 26ani.record, 26ani.replay, 25, 27animation, 25

cmdscale, 22compute.animation, 2, 3, 3, 10, 11, 18, 19,

21, 25, 27, 30, 31, 33, 38, 39

effect.edgeAgeColor (effectFun), 6effect.vertexAgeColor (effectFun), 6effectFun, 6effects (effectFun), 6export.dot, 7, 20export.pajek.net, 9

filmstrip, 2, 10

get.networks, 23

htmlwidgets, 33

install.ffmpeg, 11install.graphviz, 12, 20isoMDS, 22

layout.center, 13layout.distance, 4, 5, 8, 14, 19, 21, 22layout.normalize (layout.center), 13lines, 36

msm.sim, 2, 15, 41

ndtv, 5ndtv (ndtv-package), 2ndtv-package, 2ndtvAnimationWidget, 17ndtvAnimationWidgetOutput

(ndtvAnimationWidget), 17network, 2, 42network.collapse, 22, 23network.layout.animate, 18network.layout.animate.Graphviz, 4, 12,

13, 23network.layout.animate.kamadakawai, 4network.layout.animate.MDSJ, 4, 5, 23network.layout.animate.useAttribute, 4network.layout.kamadakawai, 19, 21networkDynamic, 2, 3, 10, 16, 30, 38

par, 10, 25, 30, 31, 36plot, 31, 32, 43plot.default, 36plot.network, 10, 21, 25–27, 31, 36–39, 43plot.network.default, 26, 31plotArgs.network, 36proximity.timeline, 2, 21, 43

read.paj, 9render.animation, 2, 3, 5, 6, 10, 11, 22, 25,

29, 31, 33render.d3movie, 2, 17, 18, 27, 29renderNdtvAnimationWidget

(ndtvAnimationWidget), 17

sammon, 22saveVideo, 25, 26scatterplot3d, 38, 39short.stergm.sim, 2, 41short.stergm.sim (stergm.sim.1), 34slice.par (compute.animation), 3stergm.sim.1, 3, 34

text, 36

45

46 INDEX

timeline, 2, 24, 35timePrism, 37toy_epi_sim, 3, 40transmissionTimeline, 42

xspline, 22, 39